Feature extraction code requirement

[L4zyy]

Thank you for sharing this great work!

I am trying to train VTimeLLM model on my own small dataset through all three stages, but I cannot find any script or code about the feat extraction prerocess. I wonder if you can kindly share some code or scirpt to extract the image feat?

[huangb23]

You can find the relevant code for feature extraction in vtimellm/inference.py. You can consider adding multi-processing or multi-GPU inference for acceleration based on your requirements.